[brlewis]

On twitter, the replies that deconstruct or fortify an argument are interspersed with a lot of junk. I've built https://en.howtruthful.com/ as a tool specifically for deconstructing and fortifying arguments. I see a lot of potential to succeed where twitter failed.

[passion__desire]

"The test of a first-rate intelligence is the ability to hold two opposed ideas in the mind at the same time, and still retain the ability to function" — F. Scott Fitzgerald.

Infact we can generalize this to N ideas cases.

Take two ideas. A and B. Give weight ∆ and 1-∆ to them. Multiple the positions of those ideas with the weights. This gives you a new interpolated position. Now repeat this with new position and idea C. And so on. You will quickly cover the area of convex hull of these ideas.

I think with future advanced LLMs the above scenario is achievable.